Evidence generated by us and others in animals provides evidence that inflammatory processes contribute to the development of at least of the early stages of diabetic retinopathy, and especially the degeneration of retinal capillaries. How this capillary degeneration in diabetes is not clear, but most studies to date have focused on metabolic defects within cells of the retina and its vasculature. The present grant proposal is based on novel observations we have made over the previous several years suggesting that proinflammatory pathways within cells derived from bone marrow are critical in the development of the retinopathy. We have found that the vascular lesions of diabetic retinopathy can be prevented merely by eliminating either iNOS or PARP-1 from the marrow-derived cells only. Which myeloid cells are responsible for development of lesions of diabetic retinopathy, how they cause the diabetes-induced abnormalities in structure and function, and how to inhibit these leukocyte-driven abnormalities are not known, and are the subjects of the proposed research. In this proposal, we will (1) characterize which marrow-derived cell types participate in the development of early stages of diabetic retinopathy, (2) determine the role of chemoattractants in recruitment of marrow-derived cells to the retina in diabetes, and if the chemoattraction is critical to diabetes-induced degeneration of retinal capillaries, and (3) investigate mechanism by which marrow-derived cells kill retinal cells in diabetes, focusing initially on the role of reactive oxygen species generated by NADPH oxidase in the development of diabetes-induced inflammation, vascular dysfunction and degeneration. We also will determine if the abnormalities within the marrow-derived cells cause degeneration of neuroglial cells of the retina. We postulate that these diabetes-induced degenerative changes in the retina can be inhibited by blocking the abnormal metabolism within the marrow-derived cells, or by blocking the adhesion between those blood cells and retinal vascular cells. This data is novel, and indicates that views of diabetic retinopathy need to expand beyond the traditional retina- or vascular-specific view of pathogenesis to include also bone marrow-derived cells.